Signature Verification Using Static and Dynamic Features

Abstract. A signature verification algorithm based on static and dynamic features of online signature data is presented. Texture and topological features are the static features of a signature image whereas the digital tablet captures in real-time the pressure values, breakpoints, and the time taken to create a signature. 1D - log Gabor wavelet and Euler numbers are used to analyze the textural and topological features of the signature respectively. A multi-classifier decision algorithm combines the results obtained from three feature sets to attain an accuracy of 98.18%.

• Type: Conference
• Year: 2004
• Where: ICONIP
• Authors: Mayank Vatsa, Richa Singh, Pabitra Mitra, Afzel Noore
• Keywords: ICONIP 2004, ICONIP 2007, Online Signature Data, Signature, Signature Verification Algorithm